Transliteration, spelling, and pronunciation

Happy New Amended Gregorian calendar!  I would say that my New Year’s resolution is to update more often, but that would be a total lie. Actually, my resolution/plan for this year is basically the PhD comic one, so updates will probably remain sporadic at best.

However, that is of the future.  The now is a really cool question from Uzza:

“moslem” as I understand it is an older spelling, contra “muslim”, which has become more popular in recent years. Without any certainty, I attribute this to the input of actual Islamic people in American discourse.  

Now some are claiming that the latter spelling is racist & etc., leaving me rather nonplussed.

Should we avoid writing “moslem”, so as not to offend? relegate it to the same wordbin as “mohammedan”?

Also, How should we pronounce it?  The first syllable  as moo or muh?  Should we say ALL-ah or al-LAH?  IS-lam, or is-LOM?

My perspective is that of a linguist, I realize these are different correct(ish) transliterations from Arabic script; I’ve not seen any accurate phonetic transcriptions, and from what I hear it’s usually pronounced like “moo- slum”, for either spelling.

Which brings up a point, one of the basic rules of English phonology is that intervocalic /z/ is voiceless. That puts the kibosh on saying moo-s-lim. Sorta. Same problem arises with the accents mentioned above, ….aaaand …we get into issues of culture, and imperialism, and holy wars,,,, and maybe I should just shut up.  But I’m curious, will people be offended. 

If you’re still reading, thanks for your persistence.  If you answer, I’ll worship you.   

–Uzza  (the Nabatean one, not that doofus from the bible)


I’m going to talk a little bit about Arabic linguistics and the history of the language’s use in the West, and then I’ll answer your question more directly.  For that bit, skip to the end!

A little bit of history

Okay, so a little background for people who haven’t studied Arabic or linguistics – Arabic is a Semitic language, like Hebrew and Aramaic, and is based on triliteral roots, meaning that most natural (non-loan) Arabic words have three root letters that are expanded and given prefixes and suffixes to be declined, conjugated and made into new words.  The standard model for this is the letters k-t-b.  “Kataba” means “to write.”  “Kitab” means “a book.”  “Katib” means “a writer.”  “Iktaba” means “to send” (as in a letter).

The Arabic alphabet has three short vowels, generally transliterated in modern Arabic transliterations (more on that below) as a, i and u.  It also has three long vowels, usually transliterated as ā, ī, and ū.  The short vowels aren’t letters, but diacritical marks that can be written in above or below the line, but aren’t required, and are generally left out, especially in handwriting.  The long vowels are letters whose pronunciation can change depending on their nearest consonant neighbor in the word or by the use of a hamza (an aspiration mark, usually transliterated with an apostrophe).

So far, so good.  But what I’ve just laid out is roughly an explanation of the modern rules of transliteration.  Transliteration of Arabic into Western languages has been around for a millenium and a half, and has rarely been systematized.   Much of the early translations and transliterations were into Latin, often by way of Greek, and were generally done phonetically.  Elements of these transliterations still exist in English today – ‘algebra’, for example, comes from the Arabic “al-jabr” meaning “the putting together of parts.”  Medieval Latin conflated the prefixed article ‘al’ as part of the word, transliterated the “j” sounds as a soft “g”, and added a vowel at the end, probably either for pronunciation or to give the word a gender.

Phonetic transliterations are very common, especially when there is only limited knowledge of the language.  Again, “algebra” is a useful example – early Latin (or possibly even Greek) translators of Arabic mathematical works didn’t consistently recognize the use of the article, and so blended it into the word itself, transliterating the word as a unique term rather than translating it.  Eventually, that transliterated term took on a unique meaning, and thus made its way into other European languages.

Phonetic transliterations remained the norm even as Semitic and Near Eastern studies developed into distinct academic fields in the eighteenth and nineteenth century, and it’s in this period that the slightly more standardized forms of “Moslem,” “Mohammed,” or “Koran” start to appear.  These words were transliterated phonetically into English and German (thus the “k” in Qur’an, where English could have just as well used a ‘c’ or ‘qu’).  Phonetic transliteration into French also gave us the “dj” seen in Djibouti and djinn, as to French ears, the Arabic “j” sound was more guttural than the French “j,” so they prefixed it with a “d” to get the sound closer.


The academization of writings about the Near East and Islam lent a certain amount of consistency to these transliterations, with “Moslem” and “Koran” emerging as the most common spellings (although English and German scholars often varied in how they capitalized them in English writing).  The term “Mohammedan” was also common, but, as Uzza rightly points out, has been effectively cast aside as being pejorative, for implying that Muslims worship Muhammad (peace be upon him) as Christians worship Christ (although I’m sad to say, I have heard it used in academic settings in the 21st century…).

Interest in standardizing Arabic transliterations developed in the late nineteenth and early twentieth century with the increase in Near Eastern and Semitic-language journals.  Printing non-Western characters was expensive, and depending on the printing style, sometimes impossible – non-Western characters had to be drawn in by hand in each copy – so printing transliterations consistently, so that the work could be disseminated more broadly and the readers could get a clear idea of the text, became a bigger priority.

It also led to a disagreement in how transliterations should work.  The other Semitic languages, especially the Aramaic dialects, which were considered dead languages, were transliterated systematically, so that each character in the Aramaic or Syriac alphabet corresponded to one (or a marked set of two, such as with ‘th’ and ‘sh’) in the English alphabet.  Although they mostly matched up in terms of sounds, the matching didn’t work 100%.  But, Syriacists argued, systematic transliteration was much more consistent, and made it easier for readers to reconstruct the text based on the transliteration.  

A similar, systematic transliteration system developed for Arabic, called the “DMG” style (for the Deutsche Morgenlaendische Gesellschaft, one of the largest Middle Eastern journals of the first half of the 20th century).  It used only three short vowels, and although largely phonetic, it had several non-phonetic matches.  For example, Arabic has three “th” sound letters.  The tha (which is pronounced like “that”) was transliterated “th,” the dhal (pronounced like “thought”) as “dh” and the dtha (that’s the old spelling, it’s pronounced like “those”) as a z with a dot under it, a special character that’s still not common in most typefaces (including google docs, where I’m typing this now!).  The z-character is not phonetically correct – at least for fusha, standard Arabic.  As it happens, some dialects do pronounce that letter as closer to a z-sound, so in that way, it was correct, but it didn’t follow ‘proper’ Arabic pronunciation.

Most modern transliteration systems are systematic, but now, due to imperial influence, many pre-systematized, phonetic transliterations are common in the Middle East and particularly in South Asia, as English in South Asia was institutionalized during the eighteenth and nineteenth centuries.

What’s right, what’s wrong, and what’s written

So how ‘should’ we say these words?  

If you want to talk like you speak fusha, then it’s “MOO-slem” (although it’s a gentle accent on the first syllable), “iss-LAM” (with a hard ‘a’, like the first ‘a’ in ‘animal’), and “ah-LAH” (with a soft ‘a’ in both syllables, a bit like a nice, relaxing ‘aaah’, and with a breathed pronunciation of the final ‘h’).  But there are hundreds of different Arabic dialects, and they all sound a little (or in some case, really, really, really) different.  Even the most hard-core Classical Arabic scholar, after spending a few months in Cairo, tends to come back dropping ‘q’s and ‘w’s all over the place (and man, those things are a pain to get out of the carpet!).

It gets even less clear when you’re talking about how we should pronounce the English word “Islam.”  Because it is an English word.  In Arabic, it’s spelled alif-sin-lam-mim.  In English, it’s spelled i-s-l-a-m, and the way it’s pronounced is going to reflect the resident language.  It’s the same reason we call Mexico “meks-i-co” and not “me-hi-co,” Hawaii “ha-wa-ii” and not “ha-va’-ii,” and Detroit “de-troit” and not “de-tway.”  And for the record, it’s also why Arabic-speakers say “am-ree-ka,” not “America,” and “bri-taw-nia” and not “Britain.”  Loanwords bend to fit the alphabet, morphology, and phonetics of their adoptive language.

Personally, I like systematic transliteration, but if we’re going to use them consistently, we would need to expand standard typesettings.  Also coming up with easier hot-keys for them would be nice.  I use full transliterations in my academic writings, but I usually drop the diacritical marks in my blogging and more ordinary writing (so I tend to write Islam and not Islām), mostly because constantly adding special characters is a pain, and I don’t use enough Arabic words in my blogging to risk a confusion over words with similar spellings but different vowels.  I also generally encourage my students to use “Muslim” and “Qur’an” over “Moslem” and “Koran” on the basis that they’re more recent and more broadly accepted in the field.

Is “Moslem” or “Koran” offensive?  That you’d have to ask a Muslim.  They’re not pejorative the way “Mohammedan” is – they’re not misrepresenting or making false claims about the culture through their use of terminology.  However, they are more Anglicized/Romanized, in particular “Koran” for “Qur’an,” as (I’ve learned in a decade of study) many Westerns seem weirded out by the use of apostrophes as diacritic marks.

In general, I support Westerns in general and Americans in particular (and white Americans in the very particular) getting more used to non-Romanized languages.  If you can pronounce John and Larry, you can pronounce Xin and Amr.  They are English words, you just need to learn how they’re pronounced, in the same way everyone has to learn that the “p” in pneumonia is silent, and that “enough” is pronounced like someone got drunk and made up a pronunciation for it.

So I guess my answer is I don’t think it’s offensive, but I also see no reason not to use the systematic spellings, since both exist.  If we’re going to champion one, I think we should champion that one.  But possibly people have strong allegiances to “Moslem.”  Anyone?  Bueller?


About askanislamicist

I'm an academic who specializes in early Islamic history and the history of religious interactions, who, in her free time, enjoys shouting into the internet.
This entry was posted in Uncategorized and tagged , , , , , . Bookmark the permalink.

3 Responses to Transliteration, spelling, and pronunciation

  1. uzza says:

    That was excellent, and now I know about enough! 😉

  2. Uzza says:

    Thank you for this. It’s an excellent response, and now I know about enough 😉

  3. Pingback: Djinn and Sidhe | askanislamicist

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s